CtrlK
BlogDocsLog inGet started
Tessl Logo

scvi-tools

Deep learning for single-cell analysis using scvi-tools. This skill should be used when users need (1) data integration and batch correction with scVI/scANVI, (2) ATAC-seq analysis with PeakVI, (3) CITE-seq multi-modal analysis with totalVI, (4) multiome RNA+ATAC analysis with MultiVI, (5) spatial transcriptomics deconvolution with DestVI, (6) label transfer and reference mapping with scANVI/scArches, (7) RNA velocity with veloVI, or (8) any deep learning-based single-cell method. Triggers include mentions of scVI, scANVI, totalVI, PeakVI, MultiVI, DestVI, veloVI, sysVI, scArches, variational autoencoder, VAE, batch correction, data integration, multi-modal, CITE-seq, multiome, reference mapping, latent space.

75

Quality

94%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

86%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a well-structured, highly actionable overview that routes to a real, well-organized bundle of reference files and scripts. Its main weakness is mild redundancy between the body's trigger/model lists and the description, and slightly implicit result-verification checkpoints.

Suggestions

Trim the 'When to Use This Skill' section or fold it into the Model Selection Guide to reduce overlap with the trigger list already present in the frontmatter description.

Add an explicit post-training verification checkpoint (e.g., inspect latent-space mixing or convergence metrics before clustering) to close the validation loop in the example workflow.

Consider moving the Model Selection Guide table to a reference file if it grows, keeping SKILL.md focused on the decision tree and workflow entry points.

DimensionReasoningScore

Conciseness

Largely efficient with tables and tight code snippets and no concept-explaining fluff, but the 'When to Use This Skill' section and the Model Selection Guide partly duplicate the capability list already in the description, so a little could be trimmed.

4 / 5

Actionability

Provides copy-paste-ready script invocations with full argument examples, a complete five-step worked workflow, executable critical-requirement snippets, and a decision tree routing to concrete reference files.

5 / 5

Workflow Clarity

Clear multi-step sequencing with an explicit data-validation checkpoint (validate_adata.py as step 1) and a decision tree, but integration-result verification (e.g., checking latent-space quality before clustering) is only implied, leaving a minor checkpoint gap.

4 / 5

Progressive Disclosure

SKILL.md is a lean overview pointing to 12 real one-level-deep reference files and 8 real scripts, all organized in navigable tables with explicit paths; all referenced bundle files exist on disk.

5 / 5

Total

18

/

20

Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is exemplary: it concisely states what the skill does, enumerates concrete capabilities, and provides an explicit, exhaustive trigger list in third person. It clearly answers both 'what' and 'when' with minimal conflict risk.

DimensionReasoningScore

Specificity

Lists eight concrete numbered actions (data integration/batch correction, ATAC analysis, CITE-seq, multiome, spatial deconvolution, label transfer, RNA velocity, general DL methods), giving comprehensive coverage of the framework's capabilities.

5 / 5

Completeness

Explicitly answers 'what' ('Deep learning for single-cell analysis using scvi-tools' plus eight enumerated capabilities) and 'when' ('This skill should be used when users need...' and 'Triggers include mentions of...') in third person, with concrete trigger phrases.

5 / 5

Trigger Term Quality

Comprehensive natural-term coverage including model names (scVI, scANVI, totalVI, PeakVI, MultiVI, DestVI, veloVI, sysVI, scArches) plus synonyms and concepts users actually say ('batch correction', 'data integration', 'multi-modal', 'CITE-seq', 'multiome', 'reference mapping', 'VAE').

5 / 5

Distinctiveness Conflict Risk

Occupies a clear niche (scvi-tools deep learning for single-cell genomics) with distinct model-name triggers, making overlap with unrelated skills minimal.

5 / 5

Total

20

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
anthropics/knowledge-work-plugins
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.